Fix reading mmCIF file where model is defined but model 1 is missing. Version bump.

fix example for new interface

CMakeLists.txt

@@ -25,7 +25,7 @@
 cmake_minimum_required(VERSION 3.16)
 
 # set the project name
-project(cifpp VERSION 2.0.1 LANGUAGES CXX)
+project(cifpp VERSION 2.0.3 LANGUAGES CXX)
 
 list(PREPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/cmake")
 
@@ -87,14 +87,14 @@ endif()
 
 # When CCP4 is sourced in the environment, we can recreate the symmetry operations table
 if(EXISTS "${CCP4}")
-	if(RECREATE_SYMOP_DATA AND NOT EXISTS "${CLIBD}/syminfo.lib")
+	if(CIFPP_RECREATE_SYMOP_DATA AND NOT EXISTS "${CLIBD}/syminfo.lib")
 		message(WARNING "Symop data table recreation requested, but file syminfo.lib was not found in ${CLIBD}")
-		set(RECREATE_SYMOP_DATA OFF)
+		set(CIFPP_RECREATE_SYMOP_DATA OFF)
 	else()
-		option(RECREATE_SYMOP_DATA "Recreate SymOp data table in case it is out of date" ON)
+		option(CIFPP_RECREATE_SYMOP_DATA "Recreate SymOp data table in case it is out of date" ON)
 	endif()
 else()
-	set(RECREATE_SYMOP_DATA OFF)
+	set(CIFPP_RECREATE_SYMOP_DATA OFF)
 	message("Not trying to recreate SymOpTable_data.hpp since CCP4 is not defined")
 endif()
 
@@ -154,16 +154,16 @@ else()
 	find_program(MRC mrc)
 
 	if(MRC)
-		option(USE_RSRC "Use mrc to create resources" ON)
+		option(CIFPP_USE_RSRC "Use mrc to create resources" ON)
 	else()
 		message("Using resources not possible since mrc was not found")
 	endif()
 
-	if(USE_RSRC STREQUAL "ON")
-		set(USE_RSRC 1)
+	if(CIFPP_USE_RSRC STREQUAL "ON")
+		set(CIFPP_USE_RSRC 1)
 
 		message("Using resources compiled with ${MRC}")
-		add_compile_definitions(USE_RSRC)
+		add_compile_definitions(CIFPP_USE_RSRC)
 	endif()
 endif()
 
@@ -204,7 +204,7 @@ string(TIMESTAMP BUILD_DATE_TIME "%Y-%m-%dT%H:%M:%SZ" UTC)
 configure_file("${CMAKE_SOURCE_DIR}/src/revision.hpp.in" "${CMAKE_BINARY_DIR}/revision.hpp" @ONLY)
 
 # SymOp data table
-if(RECREATE_SYMOP_DATA)
+if(CIFPP_RECREATE_SYMOP_DATA)
 	# The tool to create the table
 
 	add_executable(symop-map-generator "${CMAKE_SOURCE_DIR}/tools/symop-map-generator.cpp")
@@ -287,8 +287,8 @@ if (CMAKE_CXX_COMPILER_ID STREQUAL "AppleClang")
     target_link_options(cifpp PRIVATE -undefined dynamic_lookup)
 endif (CMAKE_CXX_COMPILER_ID STREQUAL "AppleClang")
 
-option(DOWNLOAD_CCD "Download the CCD file components.cif during installation" OFF)
-if(DOWNLOAD_CCD)
+option(CIFPP_DOWNLOAD_CCD "Download the CCD file components.cif during installation" OFF)
+if(CIFPP_DOWNLOAD_CCD)
 	# download the components.cif file from CCD
 	set(COMPONENTS_CIF ${PROJECT_SOURCE_DIR}/data/components.cif)
 
@@ -316,13 +316,9 @@ if(DOWNLOAD_CCD)
 endif()
 
 if(UNIX)
-	option(INSTALL_UPDATE_SCRIPT "Install the script to update CCD and dictionary files" OFF)
-	if(INSTALL_UPDATE_SCRIPT)
-		set(CIFPP_CACHE_DIR "/var/cache/libcifpp")
-		if(NOT "${CIFPP_CACHE_DIR}" STREQUAL "OFF")
-			target_compile_definitions(cifpp PUBLIC CACHE_DIR="${CIFPP_CACHE_DIR}")
-		endif()
-	endif()
+	option(CIFPP_INSTALL_UPDATE_SCRIPT "Install the script to update CCD and dictionary files" OFF)
+	set(CIFPP_CACHE_DIR "/var/cache/libcifpp" CACHE STRING "The cache directory to use")
+	target_compile_definitions(cifpp PUBLIC CACHE_DIR="${CIFPP_CACHE_DIR}")
 endif()
 
 generate_export_header(cifpp
@@ -332,7 +328,9 @@ set(INCLUDE_INSTALL_DIR ${CMAKE_INSTALL_INCLUDEDIR} )
 set(LIBRARY_INSTALL_DIR ${CMAKE_INSTALL_LIBDIR} )
 set(SHARE_INSTALL_DIR ${CMAKE_INSTALL_DATADIR}/libcifpp)
 
-target_compile_definitions(cifpp PUBLIC DATA_DIR="${CMAKE_INSTALL_PREFIX}/${SHARE_INSTALL_DIR}")
+set(CIFPP_DATA_DIR "${CMAKE_INSTALL_PREFIX}/${SHARE_INSTALL_DIR}" CACHE STRING "The directory containing the provided data files")
+
+target_compile_definitions(cifpp PUBLIC DATA_DIR="${CIFPP_DATA_DIR}")
 
 # Install rules
 
@@ -364,7 +362,7 @@ install(
 install(FILES
 	${PROJECT_SOURCE_DIR}/rsrc/mmcif_ddl.dic
 	${PROJECT_SOURCE_DIR}/rsrc/mmcif_pdbx_v50.dic
-	${PROJECT_SOURCE_DIR}/data/components.cif
+	${COMPONENTS_CIF}
 	DESTINATION ${SHARE_INSTALL_DIR}
 )
 
@@ -416,7 +414,7 @@ option(CIFPP_BUILD_TESTS "Build test exectuables" OFF)
 
 if(CIFPP_BUILD_TESTS)
 
-	if(USE_RSRC)
+	if(CIFPP_USE_RSRC)
 		add_custom_command(OUTPUT ${CMAKE_CURRENT_BINARY_DIR}/cifpp_test_rsrc.obj
 			COMMAND ${MRC} -o ${CMAKE_CURRENT_BINARY_DIR}/cifpp_test_rsrc.obj ${CMAKE_SOURCE_DIR}/rsrc/mmcif_pdbx_v50.dic ${COFF_SPEC}
 		)
@@ -463,7 +461,7 @@ message("Will install in ${CMAKE_INSTALL_PREFIX}")
 
 # Optionally install the update scripts for CCD and dictionary files
 
-if(INSTALL_UPDATE_SCRIPT)
+if(CIFPP_INSTALL_UPDATE_SCRIPT)
 	set(CIFPP_CRON_DIR "$ENV{DESTDIR}/etc/cron.weekly")
 
 	configure_file(${CMAKE_SOURCE_DIR}/tools/update-libcifpp-data.in update-libcifpp-data @ONLY)

changelog

@@ -1,3 +1,14 @@
+Version 2.0.3
+- Fixed reading mmCIF files where model numbers are used and
+  model number 1 is missing.
+
+Version 2.0.2
+- Added configuration flag to disable downloading CCD data during build
+  Note that there are now two flags for CCD data:
+  DOWNLOAD_CCD to enable downloading during build
+  INSTALL_UPDATE_SCRIPT to install an update mechanism for this file
+- Updated unit tests to work even if no CCD data is available
+
 Version 2.0.1
 - Fixed the generator for the symmetry operator table
 

examples/example.cpp

@@ -22,9 +22,7 @@ int main()
 		<< "residues with an OXT are:" << std::endl;
 	
 	for (const auto& [asym, comp, seqnr]: db.find<std::string,std::string,int>(
-			cif::Key("label_atom_id") == "OXT",
-			{ "label_asym_id", "label_comp_id", "label_seq_id" }
-		))
+			cif::Key("label_atom_id") == "OXT", "label_asym_id", "label_comp_id", "label_seq_id"))
 	{
 		std::cout << asym << ' ' << comp << ' ' << seqnr << std::endl;
 	}

include/cif++/Structure.hpp

@@ -533,6 +533,8 @@ class Structure
 	void loadData();
 	void updateAtomIndex();
 
+	void loadAtomsForModel(StructureOpenOptions options);
+
 	File &mFile;
 	size_t mModelNr;
 	AtomView mAtoms;

libcifpp.pc.in

@@ -8,5 +8,5 @@ Name: libcifpp
 Description: C++ library for the manipulation of mmCIF files.
 Version: @PACKAGE_VERSION@
 
-Libs: -L${libdir} -lcifpp @PRIVATE_LIBS@
-Cflags: -I${includedir} @PRIVATE_INC_DIRS@
+Libs: -L${libdir} -lcifpp -lboost_regex -lboost_iostreams
+Cflags: -I${includedir} -pthread

src/CifUtils.cpp

@@ -1243,6 +1243,9 @@ void addDataDirectory(std::filesystem::path dataDir)
 
 void addFileResource(const std::string &name, std::filesystem::path dataFile)
 {
+	if (not fs::exists(dataFile))
+		throw std::runtime_error("Attempt to add a file resource for " + name + " that does not exist: " + dataFile.string());
+
 	gLocalResources[name] = dataFile;
 }
 

src/Compound.cpp

@@ -636,17 +636,18 @@ void CompoundFactory::init(bool useThreadLocalInstanceOnly)
 CompoundFactory::CompoundFactory()
 	: mImpl(nullptr)
 {
+	auto ccd = cif::loadResource("components.cif");
+	if (ccd)
+		mImpl.reset(new CCDCompoundFactoryImpl(mImpl));
+	else if (cif::VERBOSE)
+		std::cerr << "CCD components.cif file was not found" << std::endl;
+
 	const char *clibd_mon = getenv("CLIBD_MON");
 	if (clibd_mon != nullptr and fs::is_directory(clibd_mon))
 		mImpl.reset(new CCP4CompoundFactoryImpl(clibd_mon));
 	else if (cif::VERBOSE)
 		std::cerr << "CCP4 monomers library not found, CLIBD_MON is not defined" << std::endl;
 
-	auto ccd = cif::loadResource("components.cif");
-	if (ccd)
-		mImpl.reset(new CCDCompoundFactoryImpl(mImpl));
-	else if (cif::VERBOSE)
-		std::cerr << "CCD components.cif file was not found" << std::endl;
 }
 
 CompoundFactory::~CompoundFactory()

src/Structure.cpp

@@ -1790,6 +1790,33 @@ Structure::Structure(File &f, size_t modelNr, StructureOpenOptions options)
 
 	auto &atomCat = (*db)["atom_site"];
 
+	loadAtomsForModel(options);
+
+	// Check to see if we should actually load another model?
+	if (mAtoms.empty() and mModelNr == 1)
+	{
+		std::optional<size_t> model_nr;
+		cif::tie(model_nr) = atomCat.front().get("pdbx_PDB_model_num");
+		if (model_nr and *model_nr != mModelNr)
+		{
+			if (cif::VERBOSE)
+				std::cerr << "No atoms loaded for model 1, trying model " << *model_nr << std::endl;
+			mModelNr = *model_nr;
+			loadAtomsForModel(options);
+		}
+	}
+
+	if (mAtoms.empty())
+		throw std::runtime_error("No atoms loaded, refuse to continue");
+
+	loadData();
+}
+
+void Structure::loadAtomsForModel(StructureOpenOptions options)
+{
+	auto db = mFile.impl().mDb;
+	auto &atomCat = (*db)["atom_site"];
+
 	for (auto &a : atomCat)
 	{
 		std::string id, type_symbol;
@@ -1805,10 +1832,9 @@ Structure::Structure(File &f, size_t modelNr, StructureOpenOptions options)
 
 		mAtoms.emplace_back(new AtomImpl(*db, id, a));
 	}
-
-	loadData();
 }
 
+
 Structure::Structure(const Structure &s)
 	: mFile(s.mFile)
 	, mModelNr(s.mModelNr)

test/unit-test.cpp

@@ -1532,8 +1532,6 @@ BOOST_AUTO_TEST_CASE(bondmap_1)
 {
     cif::VERBOSE = 2;
 
-	cif::addFileResource("components.cif", gTestDir / ".." / "data" / "components.cif");
-
 	// sections taken from CCD compounds.cif
 	auto components = R"(
 data_ASN